Significance of Normal Lung Volume on Quantitative CT Imaging Analysis in Group 1 and Group 3 Pulmonary Hypertension

Abstract

Background

Patients with concurrent pulmonary hypertension (PH) and parenchymal lung diseases have a high risk of mortality. However, whether their outcomes are related to low normal lung volume (NLV) resulting from parenchymal lung diseases on chest high-resolution CT (HRCT) imaging remains unknown.

Research Question

Would NLV on quantitative HRCT imaging affect disease behavior in patients with PH?

Study Design and Methods

This retrospective observational study evaluated patients with physician-diagnosed group 1 and group 3 PH among 1,471 patients who underwent right heart catheterization. Using a 3-dimensional image analysis system for HRCT imaging, the percentage of NLV (–950 to –600 Hounsfield units) to the whole lungs (%NLV) was calculated. The optimal cutoff point of %NLV for predicting survival was examined using the receiver operating characteristic (ROC) curve. The Kaplan-Meier method and Cox proportional hazards regression were used to detect the association between %NLV and prognosis. Multivariable logistic regression was performed to examine the association of %NLV with response to pulmonary vasodilators.

Results

Overall, 157 patients (mean age, 53.1 ± 17.6 years; sex, n = 111 [70.7%] female patients) were included. ROC curve analysis showed that the optimal cutoff of %NLV for predicting survival was 83.2%. The patients with %NLV of ≥ 83.2% showed significantly higher 5-year survival than that of those with %NLV of < 83.2% (81.7% vs 36.6%; P < .0001). Multivariable logistic regression analysis revealed %NLV of < 83.2% as an independent prognostic factor (hazard ratio, 2.49 [95% CI, 1.14–5.44]; P = .022). Responders showed significantly higher %NLV than nonresponders (90.0 ± 5.1% vs 84.7 ± 9.2%; P = .0002). Multivariable regression analysis showed that only high %NLV predicted response (OR, 1.12 [95% CI, 1.01–1.23]; P = .016).

Interpretation

Quantitative CT imaging analysis might allow numerical quantification of the lung condition and vasodilator-treatable area beyond subjective visual assessment in patients with PH. The %NLV could be a novel predictor of prognosis and treatment response in these patients.